Research Article
The Method of Anomaly Location Data Recognition Based on Improved YOLO Algorithm
@INPROCEEDINGS{10.1007/978-3-030-94551-0_5, author={Chen-can Wang and Yan Ge and Yang Li}, title={The Method of Anomaly Location Data Recognition Based on Improved YOLO Algorithm}, proceedings={Advanced Hybrid Information Processing. 5th EAI International Conference, ADHIP 2021, Virtual Event, October 22-24, 2021, Proceedings, Part I}, proceedings_a={ADHIP}, year={2022}, month={1}, keywords={Data exception Improved Yolo algorithm Exception recognition Residual error}, doi={10.1007/978-3-030-94551-0_5} }- Chen-can Wang
Yan Ge
Yang Li
Year: 2022
The Method of Anomaly Location Data Recognition Based on Improved YOLO Algorithm
ADHIP
Springer
DOI: 10.1007/978-3-030-94551-0_5
Abstract
The existing anomaly location data recognition methods usually have poor accuracy due to the rough contour curve, so the anomaly location data recognition method is studied based on the improved YOLO algorithm. The improved YOLO algorithm is designed to judge the input and output residual error comparison in the normalization process. Based on the algorithm, the abnormal data location technology is studied, and the contour curve with low noise factor is obtained. Based on the improved YOLO algorithm, the abnormal location data recognition method is designed, and the accuracy of the method is optimized. The experimental results show that in the calculation of the first type error rate and the second type error rate, the slope of the method is gentle, and the value is small. It can be seen that the method will not produce large numerical changes under the changes of mathematical expectation and regression parameters, and can more accurately realize the anomaly location data recognition.
